Abstract
We propose a method that facilitates a process of semiautomatic FrameNet construction. The method requires Google Books NGram dataset and WordNet or another thesaurus for a particular language. We evaluated the method for Russian ngrams. Due to a huge amount of available data the method does not require sophisticated natural language processing techniques (e.g. for word sense disambiguation), and it shows a promising result.
Author supplied keywords
Cite
CITATION STYLE
Ivanov, V., Ivanov, V., & Ivanov, V. (2014). Extracting frame-like structures from google books NGram dataset. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 8856, 18–27. https://doi.org/10.1007/978-3-319-13647-9_3
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
